EGFR siRNA lipid nanocapsules efficiently transfect glioma cells in vitro.

(1)

HAL Id: hal-03178900

https://hal.univ-angers.fr/hal-03178900

Submitted on 24 Mar 2021

HAL is a multi-disciplinary open access archive for the deposit and dissemination of sci- entific research documents, whether they are pub- lished or not. The documents may come from teaching and research institutions in France or abroad, or from public or private research centers.

L’archive ouverte pluridisciplinaire HAL, est destinée au dépôt et à la diffusion de documents scientifiques de niveau recherche, publiés ou non, émanant des établissements d’enseignement et de recherche français ou étrangers, des laboratoires publics ou privés.

glioma cells in vitro.

Pauline Resnier, Stephanie David, Nolwenn Lautram, Gaëtan Delcroix, Anne Clavreul, Jean-Pierre Benoit, Catherine Passirani-Malleret

To cite this version:

Pauline Resnier, Stephanie David, Nolwenn Lautram, Gaëtan Delcroix, Anne Clavreul, et al.. EGFR

siRNA lipid nanocapsules eﬀiciently transfect glioma cells in vitro.. International Journal of Pharma-

ceutics, Elsevier, 2013, 454 (2), pp.748-55. �10.1016/j.ijpharm.2013.04.001�. �hal-03178900�

(2)

InternationalJournalofPharmaceuticsxxx (2013) xxx–xxx

ContentslistsavailableatSciVerseScienceDirect

International Journal of Pharmaceutics

jou rn a l h om ep a g e :w w w . e l s e v i e r . c o m / l o c a t e / i j p h a r m

EGFR siRNA lipid nanocapsules efﬁciently transfect glioma cells in vitro

Pauline Resnier â,b , Stephanie David ^c , Nolwenn Lautram â,b , Gaëtan J.-R. Delcroix ^d,e , Anne Clavreul â,b,f , Jean-Pierre Benoit â,b , Catherine Passirani â,b,∗

aLUNAM–Universitéd’Angers,F-49933Angers,France

bINSERMU1066–MicroetNanomédecinesBiomimétiques,Angers,France

cEA6592–Nanomédicamentsetnanosondes,UniversitéFranc¸ois-Rabelais,Tours,France

dUniversityofMiamiTissueBank,InterdisciplinaryStemCellInstitute,DepartmentofOthopaedics,UniversityofMiamiMillerSchoolofMedecine,Miami, FL,USA

eGeriatricResearch,Education,andClinicalCenterandResearchService,BruceW.CarterVeteransAffairsMedicalCenter,Miami,FL,USA

fDépartementdeNeurochirurgie,CentreHospitalierUniversitaired’Angers,F-49100Angers,France

a r t i c l e i n f o

Articlehistory:

Received18February2013

Receivedinrevisedform22March2013 Accepted2April2013

Available online xxx

Keywords:

Nanomedicine Braintumor Genetherapy CH50 U87MG

a b s t r a c t

Gliomaarethemostcommonmalignanttumorsofthecentralnervoussystemandremainassociatedwith poorprognosis,despitethecombinationofchemotherapyandradiotherapy.EGFRtargetingrepresentsan interestingstrategytotreatglioma.Indeed,ahighlevelofendothelialgrowthfactorreceptorsexpression (EGFR),involvedinthemalignancyofthetumor,hasbeenobservedinglioma.Ourstrategyconsisted inusingEGFRsiRNAentrappedintolipidnanocapsules(LNCs)viacationicliposomes.Invitroanalyses onU87MGhumangliomacellswereperformedtoevaluatefirstlythecapacityofLNCstoefficiently deliverthesiRNAandsecondlytheeffectofEGFRsiRNAtargetingonU87MGproliferation.Then,the complementproteinconsumptionwasevaluatedbyCH50assaystoverifythesuitabilityofthesiRNA LNCsforsystemicadministration.TheEGFRsiRNALNCsexhibitedanadequatesizelowerthan150nmas wellasaneutralsurfacecharge.TheIC50profiletogetherwiththe63%ofproteinextinctiondemonstrated thesignificantactionofEGFRsiRNALNCscomparedtoscrambledLNCs.Doseandtime-dependentsurvival assaysshowedadecreaseofU87MGgrowthevaluatedat38%.Finally,lowcomplementconsumption demonstratedthesuitabilityofEGFRsiRNALNCsforintravenousinjection.Inconclusion,EGFRsiRNA LNCsdemonstratedtheircapacitytoefficientlyencapsulateanddeliversiRNAintoU87MGhumanglioma cells,andwillthereforebeusableinthefutureforinvivoevaluation.

1. Introduction

Malignant brain tumors are often associated with poor prog- nosis. The low survival median, of only 15 months in the case of glioblastoma (GBM), despite the use of surgery, chemotherapy with temozolomide (Temodal

^®

) and radiotherapy, has encour- aged research for new treatment (Stupp and Roila, 2009). GBM are the most aggressive form of malignant brain tumors, and are characterized by a high infiltration capacity in healthy brain tis- sue leading to difficult tumor resection. The endothelial growth factor receptor (EGFR) has been described as an oncogene in many types of cancer, including GBM (Yang et al., 2012). The over-expression and/or amplification of EGFR protein, described in half-primary GBM cases, has been associated with abnormal

∗Correspondingauthorat:INSERMU1066,IBS-CHU,4rueLarrey,49933Angers Cedex9,France.Tel.:+33244688534;fax:+33244688546.

E-mailaddress:catherine.passirani@univ-angers.fr(C.Passirani).

proliferation and a chemo/radioresistance phenomenon (Hatanpaa et al., 2010; Watanabe et al., 1996). To combat this drug resistance, various inhibitors of EGFR, such as antibodies (Cetuximab

^®

) and more recently tyrosine kinase inhibitors (TKI) (Geﬁtinib

^®

), have been developed and tested in numerous in vitro and in vivo models.

These studies have proved the relevance of EGFR targeting in GBM with anti-proliferative properties. Moreover, its impact on glioma stem cells was also demonstrated recently (Geoerger et al., 2008;

Jin et al., 2013). However, despite numerous clinical trials, none of these inhibitors have been approved yet for clinical glioma treat- ment (Joshi et al., 2012; Taylor et al., 2012; Thiessen et al., 2010). The low efﬁciency of these candidate drugs could be explained by the speciﬁc biological environment in GBM. In fact, a resistance mecha- nism against antibodies or TKI usually appears in the form of variant protein expression, rendering the therapeutic approach ineffective.

To prevent this phenomenon, the RNAi mechanism could represent a new opportunity thanks to the use of speciﬁc siRNA tools that do not depend on protein interaction, 3D conformation, or protein variants (Fire et al., 1998). Indeed, the siRNA used in this study was

http://dx.doi.org/10.1016/j.ijpharm.2013.04.001

(3)

2 P.Resnieretal./InternationalJournalofPharmaceuticsxxx (2013) xxx–xxx

described as naturally acting through the micro RNA pathway to inhibit proteins like EGFR (Michiue et al., 2009).

Systemic delivery to the brain remains a challenge due to the presence of the blood brain barrier (BBB) that strongly limits the exchanges between blood and neuronal cells. Chemical or thera- peutic agents (such as TKI, antibodies, and siRNA) cannot cross, or only partially, this membrane by passive transport (Wang et al., 2012). For this reason, the development of synthetic carriers rep- resents a great opportunity to ameliorate the passage through the BBB, especially for hydrophilic drugs. Thanks to speciﬁc targeting, nanomedicines could improve biodistribution into the brain after systemic injection by (i) the passive strategy with enhanced perme- ability and retention (EPR) effect (Maeda, 2001); (ii) by the active strategy with ligand grafting on the surface of the delivery system, such as transferrin (Beduneau et al., 2008).

Considering the systemic nucleic acid transport, nanocarriers are necessary to protect them from the blood compartment. In fact, nucleases and the immune system can rapidly degrade siRNA. The delivery systems can enhance the half-life of siRNA into blood and favor their accumulation into targeted organs (Morille et al., 2010).

Numerous nanomedicines have been developed to encapsulate siRNA, such as liposomes, micelles, and dendrimers. The majority of these systems were made of a mixture of cationic polymers and lipid components entrapping siRNA thanks to electrostatic inter- actions. Lipid nanocapsules (LNCs) consisting of a lipid liquid core of triglycerides and a rigid shell of lecithin and polyethylene gly- col, have been developed in our laboratory (Heurtault et al., 2002).

The simple formulation process is based on phase-inversion of an emulsion. These LNCs have recently been modified to encapsulate DNA, complexed with cationic lipids forming lipoplexes, and have shown efficient in vitro and in vivo effects in different animal mod- els (David et al., 2012a,b; Morille et al., 2011). In our recent work, these delivery systems were adapted to siRNA encapsulation (David et al., 2012c). We tested these siRNA LNCs on glioma cells in order to prove the efficient delivery of siRNA into the cytoplasm and the effect of EGFR inhibition by siRNA on U87MG human glioma cells.

2. Materialsandmethods

2.1. SiRNA LNC formulation

Lipid nanocapsules (LNCs) were formulated, as described before (Heurtault et al., 2002), by mixing 20%, w/w Labrafac WL 1349 (caprylic-capric acid triglycerides, Gatefossé S.A. Saint-Priest, France), 1.5%, w/w Lipoid S75-3 (Lipoid GmbH, Ludwigshafen, Germany), 17%, w/w Kolliphor

^®

HS 15 (BASF, Ludwigshafen, Germany), 1.8%, w/w NaCl (Prolabo, Fontenay-sous-Bois, France) and 59.8%, w/w water (obtained from a Milli-Q system, Millipore, Paris, France) together under magnetic stirring. Brieﬂy, three tem- perature cycles between 60 and 95

^◦

C were performed to obtain phase inversions of the emulsion. A subsequent rapid cooling and dilution with ice cooled water (1:1.4) at the phase-inversion tem- perature (PIT) led to blank LNC formation.

To obtain siRNA LNCs, the water in the last step was replaced by lipoplexes, i.e. siRNA as EGFR siRNA (sense sequence: 5

-CACAGUGGAGCGAAUUCCU-3

; Eurogentec, Seraing, Belgium) and control (scrambled) siRNA (sense sequence:

5 -UCUACGAGGCACGAGACUU-3

; Eurogentec, Seraing, Belgium) complexed with cationic liposomes in a deﬁned charge ratio of cationic lipid charge versus anionic siRNA charge (

±

ratio). Dur- ing this work, liposome LNCs, i.e. LNCs with liposome introduction at the last step without siRNA, were also used as control vehicle.

For liposome preparation, a cationic lipid DOTAP (1.2-dioleyl-3- trimethylammoniumpropane) (Avanti

^®

Polar Lipids Inc., Alabaster, AL, USA), solubilized in chloroform, was weighted in the ratio

1/1 (M/M) with the neutral lipid DOPE (1.2-dioleyl-sn-glycero-3- phosphoethanolamine) (Avanti

^®

Polar Lipids Inc., Alabaster, AL, USA) to obtain a ﬁnal concentration of 30 mM of cationic lipid charge, considering the number of lipid charges per molecule (1 for DOTAP). After chloroform evaporation under vacuum, deion- ized water was added to hydrate the lipid ﬁlm over night at 4

^◦

C which was sonicated the day after.

2.2. Characterization

The size and Zeta potential of LNCs were measured by the Dynamic Light Scattering (DLS) method using a Malvern Zetasizer

^®

apparatus (Nano Series ZS, Malvern Instruments S.A., Worcester- shire, UK) at 25

^◦

C, in triplicate, after dilution in a ratio of 1:200 with deionized water.

Moreover, a spectrophotometric method, based on work recently described by David et al. (2012c), was used to evaluate the encapsulation efﬁciency. Brieﬂy, siRNA LNCs were mixed with chloroform and water to separate respectively hydrophilic and lipophilic components, sodium hydroxide to destabilize lipoplexes and absolute ethanol to destroy LNCs. After multiple centrifu- gations, four compartments were obtained: free siRNA, free lipoplexes (i.e. siRNA associated with liposomes), encapsulated siRNA and encapsulated lipoplexes into LNCs. To determine the concentration of siRNA, the optical density (Morille et al.) of each sample were determined at 260 nm by UV spectrophotometer (UV- 2600, Shimadzu, Noisiel, France) in triplicate conditions. These physico-chemical parameters (size, charge surface and encapsu- lation) were characterized after formulation and 1 month later to study their stability.

2.3. Cell culture

Human glioma cell line U87MG and human macrophage/

monocyte THP-1 were obtained from ATCC. U87MG cells were grown in Minimum Essential Medium Eagle (Lonza, Verviers, Belgium) supplemented with 10% of fetal bovine serum (Lonza, Verviers, Belgium), 1% of antibiotics (10 units of penicillin, 10 mg of streptomycin, 25

␮

g of amphotericin B/mL; Sigma–Aldrich, Saint Louis, USA), 1% of non-essential amino acids (Lonza) and 1% of sodium pyruvate (Lonza). THP-1 macrophages were grown in Roswell Park Memorial Institute (RPMI) 1640 medium (Lonza) supplemented with adapted substrates as described in the ATCC protocol. Cell lines were cultured according to the ATCC proto- col and maintained at 37

^◦

C in a humidiﬁed atmosphere with 5%

CO

₂

. THP-1 cell adherence and differentiation into 24-well plates was allowed by medium supplemented with 100 mM phorbol 12- myristate 13-acetate (PMA, Sigma, Saint-Quentin Fallavier, France) for 48 h.

2.4. Transfection

U87MG cells were seeded onto 24-well plates at the density of 5

×

10

⁴

cells/well and precultured overnight. Before transfection, the medium was changed to a fresh one containing no serum. Treat- ment with different LNCs (blank, liposome, scrambled and EGFR siRNA) were realized at a ﬁnal dose of 37.5–300 ng siRNA/well and removed 4 h post-transfection. These cells were subsequently cul- tured for an additional time and the medium was removed every two days.

2.5. Flow cytometry

Immunostaining was performed to evaluate the expression of

membranous EGFR protein. For this, 96 h after transfection, cells

were seeded in 96-well plate, washed twice with PBS (lonza), and

(4)

P.Resnieretal./InternationalJournalofPharmaceuticsxxx (2013) xxx–xxx 3

incubated with primary antibodies (EGFR or isotype) for 1 h at 4

^◦

C. After a second wash with PBS/fetal calf serum/azide 0.02%, sec- ondary antibody coupled with FITC (polyclonal goat antimouse Ig/FITC, goat F(ab2

), Dako, Trappes, France) was incubated for 30 min at 4

^◦

C and washed. Cell suspensions were ﬁxed with PBS/azide 0.02%/formaldehyde 1%, protected from light and con- served at 4

^◦

C. Analyses were performed with a FACScalibur ﬂow cytometer (BD Bioscience, San José, USA) in collaboration with SCCAN platform (Angers).

2.6. Survival assay

Glioma cells and macrophages were seeded onto 24-well plates at a density of 5

×

10

⁴

cells/well and pre-cultured overnight or 48 h for THP-1. Dose-dependent transfection was realized as described above to determine the IC

₅₀

and cells were subsequently cul- tured for two days. Cytotoxicity assays were performed using MTS (3-(4,5-dimethylthiazol-2-yl)-5-(3-carboxymethoxyphenyl)- 2-(4-sulfophenyl)-2Htetrazolium) (Promega, Madison, USA). For this, 100

␮

L of MTS/well were disposed and plates were incubated 2.5–4 h at 37

^◦

C in a humidiﬁed atmosphere with 5% CO

2

. The OD at 492 nm was evaluated by Mutliskan Ascent (Labsystems, Fisher Scientiﬁc, Wilmington, USA).

2.7. Proliferation study

Glioma cells were seeded onto 24-well plates at a density of 5

×

10

⁴

cells/well and pre-cultured overnight. A time-dependent transfection with the low dose (10 ng) was realized and cells were subsequently cultured for ten days and the medium was removed every two days. To evaluate the progression of the glioma cells, sphere formation was followed every day. For this, scores were deﬁned: 0 corresponding to the absence of a sphere in a well, 1: irregular apparition of small spheres, 2: regular distribution of small spheres, 3: regular medium spheres, 4: regular large spheres.

The cells were photographed at day 10 after treatment. Moreover, every two day, cells were suspended by trypsin (0.5 g porcin trypsin, 0.2 g EDTA; Sigma, Saint Louis, USA), and two washes were realized with PBS. Analyses with trypan blue staining was used to count the number of cells per well.

2.8. Complement test

Complement consumption was assessed in normal human serum (NHS) (provided by the Etablissement Francais du Sang, CHU, Angers, France) by measuring the residual hemolytic capac- ity of the complement system after contact with LNCs. The ﬁnal dilution of NHS in the mixture was 1:4 (V/V) corresponding to 100

␮

L of NHS into 400

␮

L of reactive media. The technique con- sisted in determining the amount of serum able to lyse 50% of a ﬁxed number of sensitized sheep erythrocytes with rabbit anti- sheep erythrocyte antibodies (CH50), according to the procedure described by Vonarbourg et al. (2009). Complement activation was ﬁrst expressed as a function of the surface area in order to compare different formulations. Nanoparticle surface areas were estimated using the equation: S = 3m/r , where S is the surface area [cm

²

], m is the weight [g] in 1 mL of suspension, r is the average radius [cm]

determined by DLS, and is the volumetric mass [g/cm

³

] of the nanoparticles estimated at 100 g/cm

³

.

2.9. Statistical analyses

Comparisons between the treated cells of all groups were performed either using a classical analysis of variance (one-way

ANOVA) followed by a Scheffe’s post hoc analysis. Statistical signif- icance was ascribed to a threshold p-value of 0.05.

3. Results

3.1. siRNA LNC formulation 3.1.1. Size and Zeta potential

The size and Zeta potential of siRNA LNCs were determined using a Nanosizer apparatus. Measurements were realized 1 day after formulation and 1 month after storage at 4

^◦

C to evaluate the physico-chemical stability of siRNA LNCs (Table 1). Blank LNCs exhibited a stable size, narrow distribution around 60 nm and a negative Zeta potential estimated at

−

5 mV over a 1 month period.

Liposome LNCs showed a decreasing evolution in size and an increasing surface charge after 1 month at 4

^◦

C. The Zeta potential was largely positive due to cationic lipids and their polydispersity index (PdI) that decreased strongly with time from 0.26 to 0.05.

Concerning the two different lipoplexe CR used to obtain siRNA LNCs, an inﬂuence was observed with an increase of size after for- mulation with 90 nm and 122 nm respectively for CR of 2.5 and 5 compared to blank LNCs (56 nm). Moreover, this parameter was associated to a high PdI just inferior to 0.3 (limit of monodispersity).

As observed with time for liposome LNCs, siRNA LNCs presented a high reduction of size and PdI with a gain of Zeta potential (Table 1).

3.1.2. Encapsulation efﬁciency

An adapted method based on UV spectroscopy was used to eval- uate the encapsulation efficiency of siRNA into LNCs. Briefly; four compartments of siRNA localization (free siRNA, free lipoplexes, encapsulated siRNA and encapsulated lipoplexes) were separated using an easy chloroform extraction as described above. Analy- ses of EGFR siRNA LNC encapsulation efficiency were performed, in a similar way as for size, one day and 1 month after formula- tion (Table 1). The encapsulation was estimated at 29% and 40%

for respectively CR 2.5 and 5. The stability of encapsulated doses was conﬁrmed for both CR 1 month after formulation. The major- ity of encapsulated siRNA was associated with liposomes. However, a non-negligible part of siRNA was not associated with lipo- some but was still encapsulated into LNCs as previously observed.

According to their size, Zeta potential and encapsulation stabil- ity, one day-old EGFR siRNA LNCs were used in the following experiments.

3.2. Analysis of efﬁcient transfection in vitro

Four days after EGFR siRNA LNC treatment, the EGFR pro- tein level was evaluated by ﬂow cytometry (Fig. 1). No protein extinction was observed with siRNA alone in opposition to what was observed with siRNA associated with a commercial trans- fection agent (ICAfectin

^®

, Invitrogen, Saint Aubin, France) which resulted in a 55% of inhibition with a low dose of siRNA (37.5 ng) (Fig. 1). Liposome and scrambled LNCs showed no difference of EGFR expression compared to non-treated cells. Moreover, CR 2.5 LNCs had no effect on U87MG EGFR expression until 300 ng, the maximal dose (Fig. 1A). CR 5 LNCs demonstrated a dose-dependent response with a maximal inhibition estimated at 63%, equivalent to the positive control result (Fig. 1B). Consequently, CR 5 LNCs were chosen to perform in vitro assays.

3.3. Determination of IC

₅₀

The cytotoxicity of EGFR siRNA LNCs was performed on two

cell lines: U87MG glioma and THP-1 human cells in differentiated

macrophages. EGFR siRNA had only a little impact on U87MG cell

(5)

Table1

CharacterizationofEGFRsiRNALNCs.

TypeofLNCs LiposometypeandCR Date Size(nm) PdI Zeta(mV) Encapsulationyield(%)

EGFRsiRNALNCs

DOTAP/DOPE2.5 1day 90±1 0.22±0.01 +1±1 29±9

1month 56±1 0.09±0.02 +24±5 24±3

DOTAP/DOPE5 1day 122±1 0.26±0.01 +4±1 40±12

1month 70±1 0.06±0.01 +26±1 42±7

LiposomeLNCs DOTAP/DOPE5 1day 102±10 0.26±0.01 +5±2 /

1month 58±2 0.05±0.01 +34±2 /

BlankLNCs / 1day 56±1 0.04±0.01 −6±1 /

1month 60±2 0.03±0.01 −5±1 /

CR:chargeratio,PdI:polydispersityindex(monodispersity:PdI<0.3).

viability,witha low IC₅₀ evaluatedat 40ng(Fig.

2). This effect was considered speciﬁc since the survival of differentiated THP- 1 cells was not affected by the EGFR siRNA. Toxicity of liposome and scrambled LNCs was observed at 250 and 80 ng on U87MG (Fig. 2A) and 70 and 120 ng on THP-1 cells respectively (Fig. 2B).

Following these results, a dose of 10 ng siRNA corresponding to 0.1 mg/mL of LNCs was chosen to study the growth progression for 10 days as it led to cytotoxicity (30%) compared to control conditions.

Fig.1.ExpressionofEGFRproteinbyﬂowcytometryanalysesaftertreatmentwith twoCRofEGFRsiRNALNCs.TheseanalysesbyﬂowcytometryofEGFRlevels wererealizedfourdaysaftertreatmentwithincreasingdosescomprisedbetween 37.5ngand300ngofsiRNAperwell.TheEGFRproteinlevelwasexpressedinfunc- tionofnon-treatedcells.(A)AnalysesofDOTAP/DOPECR2.5LNCs.(B)Analyses ofDOTAP/DOPECR5LNCs.StatisticalanalysesANOVAonewayposthocScheffe,

***p=0.001.

3.4. Proliferation impact

After 10 days, pictures of U87MG cells were taken (Fig. 3A). The apparition of spheres into wells was recorded as described above in the cell proliferation study. A different organization of U87MG cells was evidenced between non-treated cells (a), cells treated with liposome LNCs (b) and scrambled LNCs (c) compared to cells that received EGFR siRNA LNCs (d) (Fig. 3A). EGFR siRNA LNCs signiﬁcantly decreased the formation of U87MG spheres from Days 6 to 10 (Fig. 3B). Liposome and scrambled LNCs followed the same

Fig.2.DeterminationofIC50valuesofEGFRsiRNALNCsonhumangliomacells (U87MG)(A)anddifferentiatedmacrophages(THP-1)(B).(A)EGFRsiRNAintoLNCs ( )hadanimpactonU87MGviabilitywithIC50estimatedlowerthan40ng ofsiRNAcomparedtocontrolconditionswithIC50superiorat80ng(liposome andscrambledsiRNALNCs ).(B)Onmacrophages,EGFRsiRNA LNCsdidnotmodifythecellviability.

(6)

Fig.3.(A)PicturesofU87MGrealized10daysaftertreatmentwithalowdoseofLNCs.LargesphereswereobservedonU87MGculture(a),andafter4hincubationwith liposomeLNCs(b),andscrambledLNCs(c)contrarytothetreatmentwithEGFRsiRNALNCs(d).(B)ImpactofEGFRsiRNALNCsonU87MGspheroidformation.Spheroid formationwassigniﬁcantlyinhibitedbyEGFRsiRNA( )fromDays6to10.Liposome()andscrambled()LNCspresentedthesamebehaviorasnon-treatedcells( ).

StatisticalanalysesANOVAonewayposthocScheffe,*p=0.05,**p=0.01and***p=0.001.

progression as non-treated cells with maximal index obtained at day 10 (Fig. 3B). To confirm this result, a determination of cell numbers was realized in parallel. EGFR LNC treatment presented a significant difference with all controls at day 10 (Fig. 4). On the final day, a difference above 30% was evidenced between EGFR siRNA LNCs and all control LNCs.

3.5. Complement activation

The diameter of nanoparticles was controlled by a classical method before performing CH50 assays (Fig. 5). The analysis of

complement proteins activation was performed on liposomes, lipoplexes, blank LNCs, liposome LNCs, and EGFR siRNA LNCs to compare their behavior in contact with human serum (Fig. 5).

No activation of complement proteins was evidenced with siRNA

alone (data not shown). The very low activation proﬁle of blank

LNCs was already described due to negative surface charge and a

size of around 60 nm (Morille et al., 2010). In opposition to LNCs,

positively-charged surface delivery system such as liposome and

lipoplexes led to a high consumption of complement proteins with

higher proﬁle for lipoplexes. Thanks to the neutral charge of our

device, intermediate consumption was evidenced for loaded LNCs

(7)

Fig.4.EvolutionofU87MGcellularproliferationafteronetreatmentwith10ngof EGFRsiRNALNCs.ThisdoseallowedadecreaseofcellnumberswithEGFRsiRNA LNCs( )comparedtoothertreatmentssuchasliposomeLNCs( )and scrambledsiRNALNCs( )fromDay5toDay10.Asigniﬁcantdifferencewith non-treatedcells( )wasevidencedonDay10.StatisticalanalysesANOVA onewayposthocScheffe,**p=0.01.

(liposome and siRNA), which were close to the blank LNC proﬁle until 400–500 cm

²

.

4. Discussion

In this study, we developed a speciﬁc delivery system targeting EGFR protein by using an RNAi mechanism for glioma therapy. This nanomedicine consists of the association of DOTAP/DOPE siRNA cationic lipoplexes and lipid nanocapsules (LNCs) to combine (i) the efﬁcient electrostatic interaction of lipids and siRNA, (ii) the high transfection capacity of cationic lipids and (iii) the systemic deliv- ery of nanocapsules thanks to their stealth property (Vonarbourg

et al., 2009). DOTAP and DOPE lipids have been commonly used as transfection agents and as components in nucleic acid delivery sys- tems (Resnier et al., 2013). In this work, two different charge ratios (2.5 and 5) were studied to follow their impact on siRNA LNC formu- lation. These two formulations presented a size inferior to 150 nm, a neutral Zeta potential and an encapsulation stability adequate for a long circulating proﬁle. However, measures realized 1 month later highlighted the change of physicochemical properties of siRNA LNCs over time. Size diminution and surface charge increases could signify that LNCs were reorganized in positive smaller particles, possibly due to the large amount of cationic tensio-active molecules (Lecithin, DOTAP and DOPE). This point could also explain the dif- ference of size between siRNA LNCs CR 2.5 and 5. Twice as many lipids were introduced into the CR 5 formulation compared to CR 2.5, which could lead to larger nanoparticles. Despite this phe- nomenon, the encapsulated part of siRNA was unchanged and the high CR was more interesting because it led to higher encapsulation efﬁciency, estimated at 40% of initial quantity of siRNA, which cor- responds to a concentration of 100

␮

g of encapsulated siRNA per mL of LNCs (=300 mg of LNCs).

The protein inhibition study conﬁrmed the beneﬁt of using this high charge ratio during our formulation. Indeed, it resulted in an extinction of membranous EGFR evaluated at 63%. This high extinction percentage was most likely due to our nanomedicine composition. Indeed, lipid components of siRNA LNCs, i.e., cationic lipids and lecithin, could promote the passage through cell mem- branes by their hydrophilic polar heads. The slightly positive surface charge of siRNA LNCs is expected to favor the electrostatic interactions between nanocapsules and cell membrane to facilitate their internalization. In fact, LNCs are known to be well-internalized by cancer cells, especially for the larger ones (100 nm) by multiple, non-receptor-dependent internalization pathways such as clathrin, endocytosis and cavoelae pathways (Paillard et al., 2010). Finally, a major role of these lipids in endosomal escape has already been demonstrated, especially for DOTAP and DOPE lipids involved in two well-known mechanisms: the proton sponge effect and mem- brane destabilization (Cardarelli et al., 2012; Varkouhi et al., 2011).

Fig.5. Evaluationofcomplementproteinconsumptionbylipoplexes,liposomesandLNCs.Liposomes( )andlipoplexes( )showedahighactivationprofile comparedtoLNCformulations.BlankLNCs( )werecharacterizedbyalowactivationprofileandloadedforms(liposomeLNCs andEGFRsiRNALNCs ) presentedanintermediateprofile.

(8)

Combined with the initial properties of LNCs concerning their endo- somal escape capacities, these lipids could favor this phenomenon (Paillard et al., 2010).

In the literature, the inhibition of EGFR in glioma models was assessed by shRNA form, another RNAi strategy. ShRNAs corre- spond to an intermediate form before mature siRNA is formed, and are characterized by a hairpin loop structure. However, shRNAs are not stable, so their use is generally replaced by a DNA plas- mid encoding for these shRNAs. The siRNA LNCs showed a similar extinction of EGFR protein as the shRNA EGFR used in combination with classic transfection agents (Inaba et al., 2011). PAMAM den- drimers, used for shRNA plasmid transfection, showed after 48 h transfection an inhibition of EGFR lower than siRNA LNCs on glioma cells (35% versus 63%) (Han et al., 2010). Moreover, cell prolifera- tion was signiﬁcantly reduced by our siRNA EGFR LNCs after only 10 days of treatment. Electroporation of shRNA plasmid resulted to similar results with a 40% of cell density decrease after 6 days.

However this transfection methodology, i.e. electroporation, is not translatable to clinical situations.

The benefit of LNCs is their possible use for intravenous injection as we validated their capacity to protect siRNA from complement opsonisation by CH50 assay. In fact, the low activation profile obtained in complement activation experiments confirmed the possible capacity of siRNA LNCs to circulate a long time in blood.

This result can be explained by the beneﬁt of the nanocapsule shell that hides the positive charges of the lipids thanks to weak links with negative PEG dipoles (Vonarbourg et al., 2005), thereby con- ferring stealth property to our LNCs in opposition to lipoplexes. This great enhancement in the behavior of our LNCs is more likely due to the presence of PEG chains at the surface of LNCs, as demon- strated in our previous works on DNA LNCs (David et al., 2012a). To evaluate this, in vivo experiments should be realized to deﬁne the half-life time and biodistribution of siRNA LNCs in a murine glioma model.

Interestingly, EGFR expression has also been described to be involved in chemo and radioresistance phenomena (Watanabe et al., 1996). For this, the potential of combinatory treatments could be assessed like Chen and co-workers did with doxorubicin and c- myc siRNA (Chen et al., 2010). In the future, the impact of siRNA EGFR inhibition on sensitization of classic chemotherapeutic agent or radiation could be studied in vitro and in vivo.

5. Conclusion

Our lipid nanocapsules (LNCs) were created 10 years ago by an easy process formulation and have now been fully character- ized. Moreover, previous studies have demonstrated the possibility to encapsulate hydrophilic nucleic acids into these systems. Here, our ﬁrst aim consisted of the study of the impact of CR used and the stability depending on EGFR siRNA LNC physico-chemical parameters. Analyses of protein levels allowed us to conclude on the promising delivery of siRNA into the cytoplasm by LNCs thanks to the high degree of inhibition observed. The potential of EGFR siRNA targeting to combat glioma progression was evi- denced on cell growth and U87MG sphere development. Finally, stealth properties evaluated by CH50 assay, conﬁrmed for EGFR siRNA LNCs, the possibility of future treatment to combat glioma cell proliferation alone or as an adjuvant treatment by systemic injection.

Acknowledgments

This work was supported by the foundation “Association de Recherche contre le Cancer” and by “La Ligue contre le cancer 49 et 35”. We would like to thank the SFR “Structure Fédérative

de Recherche – Interactions cellulaires et applications thérapeu- tiques” and the active participation of SCCAN platform, especially Dr Catherine Guillet for ﬂow cytometry analyses.

References

Beduneau,A.,Hindre,F.,Clavreul,A.,Leroux,J.C.,Saulnier,P.,Benoit,J.P.,2008.Brain targetingusingnovellipidnanovectors.J.Control.Release126,44–49.

Cardarelli,F.,Pozzi,D.,Bifone,A.,Marchini,C.,Caracciolo,G.,2012. Cholesterol- dependentmacropinocytosisandendosomalescapecontrolthetransfection efﬁciencyoflipoplexesinCHOlivingcells.Mol.Pharm.9,334–340.

Chen, Y., Wu,J.J., Huang, L., 2010. Nanoparticles targeted with NGR motif deliverc-mycsiRNAanddoxorubicinforanticancertherapy.Mol.Ther.18, 828–834.

David,S.,Carmoy,N.,Resnier,P.,Denis,C.,Misery,L.,Pitard,B.,Benoit,J.P.,Passirani, C.,Montier,T.,2012a.InvivoimagingofDNAlipidnanocapsulesaftersystemic administrationinamelanomamousemodel.Int.J.Pharm.423,108–115.

David,S.,Montier,T.,Carmoy,N.,Resnier,P.,Clavreul,A.,Mevel,M.,Pitard,B.,Benoit, J.P.,Passirani,C.,2012b.TreatmentefﬁcacyofDNAlipidnanocapsulesandDNA multimodularsystemsaftersystemicadministrationinahumangliomamodel.

J.GeneMed.14,769–775.

David,S.,Resnier,P.,Guillot,A.,Pitard,B.,Benoit,J.P.,Passirani,C.,2012c.siRNALNCs –anovelplatformoflipidnanocapsulesforsystemicsiRNAadministration.Eur.

J.Pharm.Biopharm.81,448–452.

Fire,A.,Xu,S.,Montgomery,M.K.,Kostas,S.A.,Driver,S.E.,Mello,C.C.,1998.Potent andspeciﬁcgeneticinterferencebydouble-strandedRNAinCaenorhabditisele- gans.Nature391,806–811.

Geoerger,B.,Gaspar,N.,Opolon,P.,Morizet,J.,Devanz,P.,Lecluse,Y.,Valent,A., Lacroix,L.,Grill,J.,Vassal,G.,2008. EGFRtyrosinekinaseinhibitionradiosen- sitizesandinducesapoptosisinmalignantgliomaandchildhoodependymoma xenografts.Int.J.Cancer123,209–216.

Han,L.,Zhang,A.,Wang,H.,Pu,P.,Jiang,X.,Kang,C.,Chang,J.,2010. Tat-BMPs- PAMAMconjugatesenhancetherapeuticeffectofsmallinterferenceRNAon U251gliomacellsinvitroandinvivo.Hum.GeneTher.21,417–426.

Hatanpaa,K.J.,Burma,S.,Zhao,D.,Habib,A.A.,2010. Epidermalgrowthfactor receptoringlioma:signaltransduction,neuropathology,imaging,andradiore- sistance.Neoplasia12,675–684.

Heurtault,B.,Saulnier,P.,Pech,B.,Proust,J.E.,Benoit,J.P.,2002. Anovelphase inversion-basedprocessforthepreparationoflipidnanocarriers.Pharm.Res.

19,875–880.

Inaba,N.,Fujioka,K.,Saito,H.,Kimura,M.,Ikeda,K.,Inoue,Y.,Ishizawa,S.,Manome, Y.,2011.Down-regulationofEGFRprolongedcellgrowthofgliomabutdidnot increasethesensitivitytotemozolomide.AnticancerRes.31,3253–3257.

Jin,X.,Jin,X.,Sohn,Y.W.,Yin,J.,Kim,S.H.,Joshi,K.,Nam,D.H.,Nakano,I.,Kim, H.,2013. BlockadeofEGFRsignalingpromotesgliomastem-likecellinvasive- nessbyabolishingID3-mediatedinhibitionofp27(KIP1)andMMP3expression.

CancerLett.328,235–242.

Joshi,A.D.,Loilome,W.,Siu,I.M.,Tyler,B.,Gallia,G.L.,Riggins,G.J.,2012.Evaluation oftyrosinekinaseinhibitorcombinationsforglioblastomatherapy.PLoSONE7, e44372.

Maeda,H.,2001. Theenhancedpermeabilityandretention(EPR)effectintumor vasculature:thekeyroleoftumor-selectivemacromoleculardrugtargeting.

Adv.EnzymeRegul.41,189–207.

Michiue,H.,Eguchi, A.,Scadeng,M., Dowdy,S.F.,2009. Induction ofinvivo syntheticlethalRNAiresponsestotreatglioblastoma.CancerBiol.Ther.8, 2306–2313.

Morille,M.,Montier,T.,Legras,P.,Carmoy,N.,Brodin,P.,Pitard,B.,Benoit,J.P.,Passir- ani,C.,2010.Long-circulatingDNAlipidnanocapsulesasnewvectorforpassive tumortargeting.Biomaterials31,321–329.

Morille,M.,Passirani,C.,Dufort,S.,Bastiat,G.,Pitard,B.,Coll,J.L.,Benoit,J.P.,2011.

TumortransfectionaftersystemicinjectionofDNAlipidnanocapsules.Bioma- terials32,2327–2333.

Paillard,A.,Hindre,F.,Vignes-Colombeix,C.,Benoit,J.P.,Garcion,E.,2010. The importanceofendo-lysosomalescapewithlipidnanocapsulesfordrugsub- cellularbioavailability.Biomaterials31,7542–7554.

Resnier,P.,Montier,T.,Mathieu,V.,Benoit,J.-P.,Passirani,C.,2013. Currentstatus ofsiRNAdeliverysystemstocombatcancers:fromneedsandchallengesto opportunitiesandclinicalperspectives.Adv.DrugDeliv.Rev..

Stupp,R.,Roila,F.,2009. Malignantglioma:ESMOclinicalrecommendationsfor diagnosis,treatmentandfollow-up.Ann.Oncol.20,126–128.

Taylor,T.E.,Furnari,F.B.,Cavenee,W.K.,2012. TargetingEGFRfortreatmentof glioblastoma:molecularbasistoovercomeresistance.Curr.CancerDrugTargets 12,197–209.

Thiessen,B.,Stewart,C.,Tsao,M.,Kamel-Reid,S.,Schaiquevich,P.,Mason,W.,Easaw, J.,Belanger,K.,Forsyth,P.,McIntosh,L.,Eisenhauer,E.,2010.AphaseI/IItrialof GW572016(lapatinib)inrecurrentglioblastomamultiforme:clinicaloutcomes, pharmacokineticsandmolecularcorrelation.CancerChemother.Pharmacol.65, 353–361.

Varkouhi,A.K.,Scholte,M.,Storm,G.,Haisma,H.J.,2011. Endosomalescapepath- waysfordeliveryofbiologicals.J.Control.Release151,220–228.

Vonarbourg,A.,Passirani,C.,Desigaux,L.,Allard,E.,Saulnier,P.,Lambert,O.,Benoit, J.P.,Pitard,B.,2009. TheencapsulationofDNAmoleculeswithinbiomimetic lipidnanocapsules.Biomaterials30,3197–3204.

(9)

Vonarbourg,A.,Saulnier,P.,Passirani,C.,Benoit,J.P.,2005.Electrokineticproperties ofnonchargedlipidnanocapsules:inﬂuenceofthedipolardistributionatthe interface.Electrophoresis26,2066–2075.

Wang,T.,Agarwal,S.,Elmquist,W.F.,2012.Braindistributionofcediranibislimited byactiveefﬂuxattheblood-brainbarrier.J.Pharmacol.Exp.Ther.341,386–395.

Watanabe,K.,Tachibana,O.,Sata,K.,Yonekawa,Y.,Kleihues,P.,Ohgaki,H.,1996.

OverexpressionoftheEGFreceptorandp53mutationsaremutuallyexclusivein

theevolutionofprimaryandsecondaryglioblastomas.Brain.Pathol.6,21–223, discussion23–24.

Yang,W.,Xia,Y.,Cao,Y.,Zheng,Y.,Bu,W.,Zhang,L.,You,M.J.,Koh,M.Y.,Cote, G.,Aldape,K.,Li,Y.,Verma,I.M.,Chiao,P.J.,Lu,Z.,2012. EGFR-inducedand PKCepsilonmonoubiquitylation-dependentNF-kappaBactivationupregulates PKM2expressionandpromotestumorigenesis.Mol.Cell48,771–784.